Continuously monitor your network speed and quality!

$ wifi-watch
PING ns.google.com (216.239.32.10): 56 data bytes

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
16:55  300  298 good   0.7%  250.83 ms
17:00  300  296 good   1.3%  219.14 ms
17:03  136   67 good   1.5%  676.17 ms

Run wifi-watch in a terminal so that you can easily check your network quality at this moment and how it has been holding up since you began recording hours or days ago.

wifi-watch will run ping every second and refresh the status line with the current counts. At the end of an accounting period the current counts are saved as a history line and a new accounting period begins.

This can be invaluable in monitoring your home internet for systematic service degradation. Service technicians pay attention when you can give them exact ping times and packet loss history!

It is also a very helpful in public wifi environments to be able to see how the shared bandwidth is holding up on a moment to moment basis.

You can install the gem from rubygems.org

gem install wifi-watch

wifi-watch is a stand alone ruby script which will run with any ruby version and has no dependencies. You can git clone the repository and copy wifi-watch into your path.

git clone https://github.com/jgorman/wifi-watch.git
sudo cp wifi-watch/bin/wifi-watch /usr/local/bin

You can use curl or wget to download the script.

curl https://raw.githubusercontent.com/jgorman/wifi-watch/master/bin/wifi-watch >wifi-watch
chmod +x wifi-watch
sudo mv wifi-watch /usr/local/bin

-h, --host host         Host to ping [ns.google.com].
-c, --count seconds     Accounting period length [600 seconds].
-p, --periods periods   Accounting periods to run [infinite].
    --ping-to file      Write ping output to a file for debugging.
    --ping-from file    Ping input from a file ('-' for stdin).
-v, --version           Version.
    --help              This message.

The status line shows counts for the current period.

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
17:03  136   67 good   1.5%  676.17 ms

• Current time.
• 136 pings attempted so far in the current period.
• Run of 67 good pings in a row!
• Status of the last ping: "good" or "fail".
• Packet loss percentage so far.
• Round trip time of the last successful ping.

Here is the past half hour of my network quality history.

In this example things were going great, then my connection slowed down over the last 15 minutes. I am working in a crowded coffee shop right now and perhaps a few of us are watching videos.

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
16:30  300  298 good   0.7%   63.84 ms
16:35  300  300 good   0.0%   76.28 ms
16:40  300  300 good   0.0%   74.45 ms
16:45  300  297 good   1.0%  129.55 ms
16:50  300  296 good   1.3%  233.94 ms
16:55  300  298 good   0.7%  250.83 ms
17:00  300  296 good   1.3%  219.14 ms
17:03  136   67 good   1.5%  676.17 ms

• History lines show the end time for each period.
• Count of good pings over the period.
• Packet loss percentage.
• Average round trip time over the entire period.

On the status line the run length count resets to 1 on every transition between "good" and "fail" modes. This way we can tell at a glance how long we have been in the current state. In a still connected high packet loss environment the runs will be short and switch modes often.

Here we have lost our internet connection for the last 39 seconds.

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
17:59  207   39 fail  19.3%   62.55 ms

At the beginning of an accounting period ping does a DNS lookup to find your target host ip address. Sometimes this can take quite a while to either succeed or fail. If the wait is more than one second you will see a status message.

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
14:38   10    4 good  60.0%   35.69 ms
14:38:52 Waiting for dns lookup of ns.google.com ...
14:39   10   10 good   0.0%   49.52 ms

When the network is not connected at the beginning of an accounting period the host DNS lookup will fail to resolve.

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
08:40  300   51 good  83.0%   68.65 ms
08:41:06 ping: cannot resolve ns.google.com: Unknown host
08:42   62   17 fail  27.4%  187.30 ms

Sometimes your coffee shop wifi login will time out and you will need to login again. Wifi-watch will let you know by showing the response message.

10:45:43 92 bytes from 10.128.128.128: Communication prohibited by filter

It is easy to capture the raw ping output to a file for later replay and testing.

wifi-watch --count 10 --ping-to test1.log
PING ns.google.com (216.239.32.10): 56 data bytes

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
10:06   10   10 good   0.0%  127.41 ms
10:06   10    3 good  70.0%   34.28 ms
10:06:34 ping: cannot resolve ns.google.com: Unknown host
10:06    5    1 fail  20.0%   55.16 ms^C

The ping output file includes the timestamp for each line so replaying the file later on should result in identical appearing output.

wifi-watch --ping-from test1.log
PING ns.google.com (216.239.32.10): 56 data bytes

Time  Ping  Run Mode Failed Round Trip
----- ---- ---- ---- ------ ----------
10:06   10   10 good   0.0%  127.41 ms
10:06   10    3 good  70.0%   34.28 ms
10:06:34 ping: cannot resolve ns.google.com: Unknown host
10:06    5    1 fail  20.0%   55.16 ms

Although the --ping-from output should appear to be the same as live ping monitoring there is a difference. Instead of showing every status line, only the final summary line is printed. This helps keep test cases short and readable.

You can run the test suite with rake test.

Bug reports and pull requests are welcome on GitHub at https://github.com/jgorman/wifi-watch.

If you discover a shortcoming capture the ping output using --ping-to and open an issue. Thanks!

The gem is available as open source under the terms of the MIT License.